For hedge fund which one would you use Std dev or downside deviation Justify in one line your choice(S)
sdn deviation normally but if the found has a historical record of positive returns you would use downside deviation to capture the true risk of the fund and not be penialize for upside deviation, so to speak.
well if a hedge fund use deriv, which most Hfs do…it is agreed that std dev is not a good measure…if thats the case why would downside be a good measure? I am quoting a question in the CFAI material…
Std. deviation isn’t a good measure but its a case of the best we have. downside is only a good measure if the fund has a history of positive returns.
Actually I would say for hedge funds downside deviation should be used instead of standard deviation. Standard deviation does not to a good job of capturing the probability of returns in the tails. Hedge Funds can exhibit extreme positive returns in a given period. These extreme positive results can skew the dispersion (standard deviation). Investors are more concerned with downside risk and as a result, downside deviation would be a better measurement. In addition, Momentum Hedge Funds are an example of a hedge fund that exhibits serially correlated returns - and this would again result in skewed return distribution making standard deviation inappropriate. Lastly, stale pricing leads to smoothed return data and underestimated standard deviation. This is a problem in general for any return variability measurement, not just standard deviation.
PJ - can you point to the CFAI or schweser where they say downside? I believe they say SD and not downside except for the reasons i explained above. you could be correct wiht your last post, but i think the CFAI says something differently on this. I think your answer is too ‘real world’ for this exam, haha. but can you point to a part of the text that agrees with you? I think this would be a great one to have an answer on.
strikershank Wrote: ------------------------------------------------------- > sdn deviation normally but if the found has a > historical record of positive returns you would > use downside deviation to capture the true risk of > the fund and not be penialize for upside > deviation, so to speak. This is bang on… exactly what I was thinking 
We agree !! see
I’ll have to check my notes later tonight when I get home… Taking the wife out for our 9 year anniversary
Schweser Book 4 pgs 34 and 35 CFAI Volume 4 pg 334
Congrats PJ.
besides the page number et al…I think if std. dev. is not applicable because of non-normality why would downside deviation apply? downside deviation is also calculated assuming returns are normal i.e returns are symmetric?
applies because of what i said in my second reply - its the best we have right now.
Thanks BW! We had fun last night but now it’s back to studying :~(
Hahah…yeah I was sitting down watch Discovery Channel about Sasquatch and my wife goes, “dont you have to study”… DAMN HER!